Diastereoselective synthesis of piperazines by manganese-mediated reductive cyclization.

Article abstract of DOI:10.1021/ol034469l

A simple and effective synthesis of trans aryl-substituted piperazines using a Bronsted acid and manganese(0) is described. [reaction: see text]

Full text of DOI:10.1021/ol034469l

ORGANIC
LETTERS
2003
Vol. 5, No. 9
1591-1594
Diastereoselective Synthesis of
Piperazines by Manganese-Mediated
Reductive Cyclization
Gregory J. Mercer and Matthew S. Sigman*
Department of Chemistry, UniVersity of Utah, 315 S. 1400 East,
Salt Lake City, Utah 84112-8500
sigman@chem.utah.edu
Received March 18, 2003
ABSTRACT
A simple and effective synthesis of trans aryl-substituted piperazines using a Brønsted acid and manganese(0) is described.
A piperazine ring gives inherent rigidity to two hydrogen
bond participants and lends itself well to binding within
biological systems. Accordingly, piperazines are found in a
large number of biologically active compounds.1 Addition-
ally, this structural rigidity has led to the use of piperazines
as ligands in asymmetric catalysis.2 Whereas piperazines with
2,5-substitution can be synthesized via reduction of the
corresponding diketopiperazine,3 2,3-substituted piperazines
are much more difficult to access. This substitution can have
an influence on their biological activity.1f Therefore, the
development of an effective, simple method for their
synthesis is desirable. An attractive strategy for the synthesis
of 2,3-substituted piperazines is the reductive cyclization of
a bisimine (Scheme 1). The imine substrates are easy to
Scheme 1. Reductive Dimerization of Imines
access by condensation of the corresponding diamine and
carbonyl compound, and the carbon-carbon bond forming
procedure sets two stereocenters in one step. Intermolecular
reductive dimerization of imines has been previously ac-
complished to form vicinal diamines using various methods
including Mg and alkali metal,4 photolysis,5 other metal
reductants,6-14 and electrochemistry.15 Stereoselectivity has
been a significant issue, with only a few methods yielding
(1) For examples, see: (a) Tagat, J. R.; Steensma, R. W.; McCombie,
S. W.; Nazareno, D. V.; Lin, S.; Neustadt, B. R.; Cox. K.; Xu, S.; Wojcik,
L.; Murray, M. G.; Vantuno, N.; Baroudy, B. M.; Stizki, J. M. J. Med.
Chem. 2001, 44, 3343-3346. (b) Zhang, Y.; Rothman, R. B.; Dersch, C.
M.; de Costa, B. R.; Jacobson, A. E.; Rice, K. C. J. Med. Chem. 2000, 43,
4840-4849. (c) Matecka, D.; Rothman, R. B.; Radesca, L.; de Costa, B.
R.; Dersch, C. M.; Partilla, J. S.; Pert, A.; Glowa, J. R.; Wojnicki, F. H. E.;
Rice, K. C. J. Med. Chem. 1996, 39, 4704-4716. (d) Glowa, J. R.;
Fantegrossi, W. E.; Lewis, D. B.; Matecka, D.; Rice, K. C.; Rothman, R.
B. J. Med. Chem. 1996, 39, 4689-4691. (e) Corey, E. J.; Gin, D. Y.; Kania,
R. S. J. Am. Chem. Soc. 1996, 118, 9202-9203. (f) Giardina´, D.; Gulini,
U.; Massi, M.; Piloni, M. G.; Pompei, P.; Rafaiani, G.; Melchiorre, C. J.
Med. Chem. 1993, 36, 690-698. (g) Giardina´, D.; Brasili, L.; Gregori, M.;
Massi, M.; Picchio, M. T.; Quaglia, W.; Melchiorre, C. J. Med. Chem. 1989,
32, 50-55. (h) Witiak, D. T.; Trivedi, B. K. J. Med. Chem. 1981, 24, 1329-
1332. (i) Manoury, P. M.; Dumas, A. P.; Najer, H. J. Med. Chem. 1979,
22, 554-559.
(4) (a) Bachmann, W. E. J. Am. Chem. Soc. 1931, 53, 2672-2676. (b)
Smith, J. G.; Veach, C. D. Can. J. Org. Chem. 1966, 44, 2497-2502. (c)
Eisch, J. J.; Kaska, D. D.; Peterson, C. J. J. Am. Chem. Soc. 1966, 88,
453-456. (d) Smith, J. G.; Ho, I. J. Org. Chem. 1972, 37, 653-656.
(5) Beak, P.; Payet, C. R. J. Org. Chem. 1970, 35, 3281-3286.
(6) Zn: (a) Smith, J. G.; Boettger, T. J. Synth. Commun. 1981, 11, 61-
64. (b) Alexakis, A.; Aujard, I.; Kanger, T.; Mangeney, P. Org. Synth. 1999,
76, 23. (c) Tsukinoki, T.; Nagashima, S.; Mitoma, Y.; Tashiro, M. Green
Chem. 2000, 2, 117-119.
(2) Shono, T.; Kise, N.; Shirakawa, E.; Matsumoto, H.; Okazaki, E. J.
Org. Chem. 1991, 56, 3063-3067.
(3) Jung, M. E.; Rohloff, J. C. J. Org. Chem. 1985, 50, 4909-4913.
10.1021/ol034469l CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/08/2003

the more useful trans product in good yields.15,16 In most of
these cases, nontrivial imine substrates are required. In
contrast, considerably less effort has been afforded to
piperazine synthesis via reductive cyclization.17 Electro-
chemical methods are most commonly used to generate
piperazine products in moderate to good yields and diaste-
reoselectivity. A general metal-mediated method would thus
be an attractive alternative.
substitution of both the Lewis acid and silyl chloride with a
single reagent seemed prudent. It was reasoned that these
additives could be acting as Lewis acids by binding imine
substrate to facilitate reduction, by activating the manganese
surface in situ, and/or by turning over any intermediates.
Thus, a simple Brønsted acid was considered a reagent that
could potentially fulfill all of these roles.
On the basis of previous uses of Brønsted acids in pinacol
couplings,18 the investigation was initiated by treating
bisimine 3a with Mn(0) (325 mesh) and pyridinium hydro-
chloride in various solvents. It was found that reductive
cyclization occurred in most solvents with the best yield of
93% obtained in 10% toluene/acetonitrile (Table 1, entry 2).
Our approach was based on a serendipitous observation
from a related project directed at carbon-carbon bond
forming reactions of imines. We found that intermolecular
imine reductive dimerization was a significant byproduct.
After determining the necessary reagents for this transforma-
tion, dimerization product 2 was found to result from the
combination of Mn(0) metal, a silyl chloride, and a Lewis
acid (Figure 1). Optimization of this process was initially
Table 1. Screen of Solvents and Brønsted Acids
entry
Brønsted acid
pyridine-HCl
p yr id in e-HCl
pyridine-HCl
pyridine-HCl
pyridine-HCl
solventa
yield (%)
1
2
3
4
5
6
7
8
acetonitrile
83
93
46
57
67
55
26
33
80
67
95
nr
a ceton itr ileb
dichloroethane
methanol
Figure 1. Original observation.
dimethoxyethane
toluene
sought through evaluation of combinations of different Lewis
acids and silyl chlorides, with limited success. Since the
diastereoselectivity for this transformation was poor, we
sought to explore and optimize a Mn-mediated intramolecular
reductive dimerization for piperazine synthesis.
pyridine-HCl
triethylamine-HCl
triethylamine-HCl
lutidine-HCl
lutidine-HCl
tr iflu or oa cetic a cid
acetic acid
acetonitrileb
dimethoxyethane
acetonitrileb
dimethoxyethane
a ceton itr ileb
acetonitrileb
9
10
11
12
In applying this method to piperazine formation, an
experimentally simple procedure was desired. Therefore,
a The use of diethylether, DMF, DMSO, acetone, and ethyl acetate
resulted in little product formation. b 10% toluene was added to enhance
solubility of the substrate.
(7) Sm(II): (a) Annuziata, R.; Benaglia, M.; Cinquini, M.; Cozzi, F.;
Raimondi, L. Tetrahedron Lett. 1998, 39, 3333-3336. (b) Enholm, E. J.;
Forbes, D. C.; Holub, D. P. Synth. Commun. 1990, 20, 981-987. (c)
Imamoto, T.; Nishimura, S. Chem. Lett. 1990, 1141-1142. (d) Aurreco-
echea, J. M.; Ferna´ndez-Acebes, A. Tetrahedron Lett. 1992, 33, 4763-
4766.
Additionally, low substrate concentration was necessary to
avoid oligomerization. Furthermore, this process is chemose-
lective in that benzaldehyde does not undergo a pinacol
coupling under these reaction conditions.
Encouraged by the excellent yields obtained with pyridine-
HCl, other simple Brønsted acids were explored. Of these,
trifluoroacetic acid (TFA) gave a comparable success with
a yield of 95% (entry 11). It is important to note that the
acidity seemed to directly correlate with yield of the process.
Both pyridine-HCl and TFA, the most effective acids
investigated, are considerably more acidic in DMSO than
triethylamine-HCl and acetic acid.19
(8) Pb/Al: Tanaka, H.; Dhimane, H.; Fujita, H.; Ikemoto, Y.; Torii, S.
Tetrahedron Lett. 1988, 29, 3811-3814.
(9) Mn: Rieke, R. D.; Kim, S. J. Org. Chem. 1998, 63, 5235-5239.
(10) Yb: Takaki, K.; Tsubaki, Y.; Tanaka, S.; Beppu, F.; Fujiwara, Y.
Chem. Lett. 1990, 203-204.
(11) In: Kalyanam, N.; Rao, G. V. Tetrahedron Lett. 1993, 34, 1647-
1648.
(12) Ti: Talukdar, S.; Banerji, A. J. Org. Chem. 1998, 63, 3468-3470.
(13) V: Hatano, B.; Ogawa, A.; Hirao, T. J. Org. Chem. 1998, 63, 9421-
9424.
(14) Zr: Buchwald, S. L.; Watson, B. T.; Wannamaker, M. W.; Dewan,
J. C. J. Am. Chem. Soc. 1989, 111, 4486-4494.
(15) Siu, T.; Li, W.; Yudin, A. K. J. Comb. Chem. 2001, 3, 554-558.
(16) For d/l selective protocols, see: (a) Al/Sm: Yanada, R.; Okaniwa,
M.; Kaieda, A.; Ibuka, T.; Takemoto, Y. J. Org. Chem. 2001, 66, 1283-
1286. (b) Mo: Cameron, T. M.; Ortiz, C. G.; Abboud, K. A.; Boncella, J.
M.; Baker, R. T.; Scott, B. L. Chem. Commun. 2000, 573-574. (c) Zn/Cu:
Shimizu, M.; Iida, T.; Fujisawa, T. Chem. Lett. 1995, 609-610. (d) Ti:
Mangeney, P.; Tjero, T.; Alexakis, A.; Grosjean, F.; Normant, J. Synthesis
1988, 255-257. (e) Nb: Roskamp, E. J.; Pedersen, S. F. J. Am. Chem.
Soc. 1987, 109, 3152-3154.
Using the optimized conditions (either 3 equiv of pyridine-
HCl or TFA, 1.5 equiv of 325 mesh Mn(0), 0.05 M substrate
in 10% toluene/acetonitrile), the scope of the reductive
(17) (a) Kise, N.; Oike, H.; Okazaki, E.; Yoshimoto, M.; Shono, T. J.
Org. Chem. 1995, 60, 3980-3992. (b) Shono, T.; Kise, N.; Oike, H.;
Yoshimoto, M.; Okizaki, E. Tetrahedron Lett. 1992, 33, 5559-5562. (c)
Shono, T.; Kise, N. Shirakawa, E.; Matsumoto, H.; Okazaki, E. J. Org.
Chem. 1991, 56, 3063-3067. (d) Betschart, C.; Schmidt, B.; Seebach, D.
HelV. Chim. Act. 1988, 71, 1999-2021.
(18) Brønsted acids have been used in a similar manner for pinacol
couplings; see: (a) Gansa¨uer, A.; Bauer, D. Eur. J. Org. Chem. 1998, 2673-
2676. (b) Gansa¨uer, A.; Bauer, D. J. Org. Chem. 1998, 63, 2070-2071.
(19) (a) Kolthoff, Z. M.; Chantooni, M. K., Jr. J. Am. Chem. Soc. 1968,
90, 23. (b) Courtot-Coupez, J.; LeDemezet, M. Bull. Soc. Chim. Fr. 1969,
1033. (c) Bordwell, F. G.; Algrim, D. J. Org. Chem. 1976, 41, 2508.
1592
Org. Lett., Vol. 5, No. 9, 2003

Table 2. Intramolecular Reductive Dimerization Scope
Table 3. Intramolecular Reductive Dimerization of Other
Substrates
entry Brønsted acid
R
product yield (%)a
1
2
3
4
5
6
7
8
pyridine-HCl
TFA
pyridine-HCl
TFA
pyridine-HCl
TFA
Ph
Ph
4a
4a
4b
4b
4c
4c
4d
4d
4e
4e
4f
93
95
96
96
96
99
89
90
76
99
93
85
80
97
nr
2,5-(CH3)2C6H3
2,5-(CH3)2C6H3
2,4-(CH3)2C6H3
2,4-(CH3)2C6H3
pyridine-HClb p-CH3OC6H4
TFAb
p-CH3OC6H4
p-ClC6H4
p-ClC6H4
a Conditions: 1.5 equiv of Mn(0), 3 equiv of TFA, 4 h, rt, 9:1 MeCN/
toluene. b Conditions: 1.5 equiv Mn(0), 3 equiv pyridine-HCl, 4 h, rt, 9:1
MeCN/toluene. c 0.01 M substrate.
9
pyridine-HCl
TFA
10
11
12
13
14
15
16
pyridine-HClc 2-furyl
TFAc
2-furyl
4f
pyridine-HCl
TFA
pyridine-HCl
TFA
2-naphthyl
2-naphthyl
cyclohexyl
tert-butyl
4g
4g
4h
4i
4k in a lower yield. Of particular note is the diastereoselective
synthesis of the seven-membered heterocyclic ring 4l in good
yield.23
nr
a Average of at least two experiments. b 4 equiv of acid is used for
optimal yields. c 0.01 M in substrate.
The high diastereoselectivity of this process deserves
comment. Although the precise role of the Brønsted acid is
currently unknown, the process presumably involves activa-
tion of the imine by the Brønsted acid followed by reduction
of the iminium on the manganese surface to form a carbon-
centered radical. Three possible scenarios can be considered
for carbon-carbon bond formation: (1) radical addition
cyclization; (2) a second one-electron reduction to form the
carbanion, which undergoes rapid intramolecular nucleophilic
addition to the activated imine; and (3) formation of the
diradical with subsequent termination. Shono and co-work-
ers’ results from related electrochemical studies indicate that
the carbanion is more difficult to form than the diradical.2
If a radical cyclization is occurring, the piperazine product
would arise from a potentially unfavorable 6-endo cycliza-
tion. Additionally, in the case of the seven-membered ring
formation, no 6-exo product is observed. Therefore, the likely
mode of cyclization is through intramolecular termination
of a diradical (Figure 2).24 A six-membered transition state
dimerization of imines was evaluated.20 Overall, aryl-
substituted imines were excellent substrates, producing the
desired piperazines in yields ranging from 76% to 99%. Both
electron-rich (entries 3-8) and electron-poor imines (entries
9 and 10) undergo reductive cyclization as well as aryl rings
with multiple substituents (entries 3-6). Furyl- and naphthyl-
substituted imines also cleanly react to yield substituted
piperazines. Notably, only one diastereomer is observed for
all cases by NMR analysis. This single diastereomer was
confirmed to be trans both by NMR assignment and by
separation of the enantiomers by HPLC equipped with a
chiral stationary phase. Scale-up of this procedure to 10 g
of 3b resulted in a 96% yield of the piperazine product 4b.
This showcases the ability to produce laboratory scale
quantities. Although the use of either Brønsted acid leads to
similar yields, the TFA-mediated process is simpler in terms
of overall workup.20 Aliphatic substrates proved unreactive
under the optimal conditions, which is likely due to the higher
reduction potential of these substrates.21
Other substrates that yield cyclic diamines were evaluated
using the optimal conditions (Table 3). Using a diamine
derived from (R,R)-cyclohexyldiamine 3j, a high yielding,
diastereoselective reductive cyclization results.22 Only one
diastereomer is observed by NMR analysis. Diamines similar
to 4j have been used in the enantioselective addition of
diethylzinc to aldehydes.2 A ketimine 3k undergoes a
diastereoselective reductive cyclization to form piperazine
Figure 2. Model for diastereoselectivity.
model for diradical termination is consistent with the
observed diastereocontrol. In this model, the imine substit-
uents are oriented trans to each other in equatorial sites.
(20) See Supporting Information for details.
(21) Cyclic voltammetry of an aryl imine in CH3CN gave a reduction
potential of -1.51 eV, whereas an aliphatic imine did not cleanly reduce
before solvent reduction was observed (ca. -1.9 eV). See Supporting
Information for details.
(22) Stereochemistry was assigned on the basis of a previous synthesis;
see ref 2.
(23) The analogous eight-membered product is produced in significantly
lower yields (10-20%).
(24) This is a similar model to that proposed by Shono and co-workers
for electrochemically mediated reductive cyclization of imines in ref 2.
Org. Lett., Vol. 5, No. 9, 2003
1593

In summary, a simple method has been developed for the
diastereoselective synthesis of 2,3-aryl-substituted pipera-
zines. The use of both Mn(0) and a simple Bronsted acid
provides for a convenient protocol for the synthesis of this
important class of heterocycles. The method has been shown
to work for a variety of aryl substrates and can be easily
accomplished on a 10 g scale. Additionally, highly substituted
piperazines and seven-membered heterocyclic rings can be
formed in good yields. The applications of substituted
piperazines to asymmetric catalysis and investigation into
other Mn(0)-mediated processes are subjects of current and
future research efforts.
Acknowledgment. This work was supported by the
National Institutes of Health (NIGMS GM63540). We thank
University of Utah Research Foundation, Merck Research,
and Rohm and Haas Research for support of this research.
We thank Mitchell Schultz for verifying the procedure and
Brian Fitchell for help with cyclic voltammetry.
Supporting Information Available: Experimental de-
1
tails, characterization of new compounds, and H NMR
spectra for all piperazines formed. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL034469L
1594
Org. Lett., Vol. 5, No. 9, 2003